央行与非专家的沟通：来自随机实地实验的见解

Central bank communication withnon-experts: insights from arandomized field experiment Alexander Jung, Francesco Paolo Mongelli Disclaimer:Thispaper should not be reported as representing the views of the European Central Bank(ECB). The views expressed are those of the authors and do notnecessarily reflect those of the ECB. Abstract We conducted a randomized field experiment to investigate the impact of direct central bankcommunication on the monetary policy knowledge and expectations of non-experts. Between 2022 and2025, at the ECB Visitor Centre, we exposed nearly 5,000 participants from diverse visitor groups topolicy briefings, similar to the communication at ECB press conferences. The random assignment ofparticipants to treatment, placebo, or control groups created exogenous variation in the information theyreceived. We implemented a new survey to gain insights into participants’ monetary policy knowledge,as well as their inflation and growth expectations, both before and after the information treatment. Thisapproach enabled us to explore how knowledge and trust influence the economic expectations of non-experts. Our findings indicate that treated individuals improved their monetary policy knowledge anddeveloped a better understanding of the ECB’s inflation target, which proved to be a key factor inexplaining why a substantialshare of treated non-experts revised their medium-term inflationexpectations, aligning them with the inflation target. However, the improved understanding of monetarypolicy did not lead to systematic revisions in their economic growth expectations. Removing potentiallanguage barriers amplified the effects of direct communication on monetary policy knowledge andindividual inflation expectations, confirming the importance of addressing citizens in their nativelanguage. JEL Codes: C83, C93, D83, D84, E31, E58. Non-technical summary Effective communication strategies with diverse stakeholders are essential for central banks to maintainprice stabilityand ensure the smooth transmission of monetary policy measures. This research examinesthe impact of direct central bank communication with non-experts on their understanding of monetarypolicy and macroeconomic expectations. Our study is motivated by the need to understand better howeffectively such communication conveys information to the public, enabling them to form well-informedeconomic expectations. A key challenge stems from the often found dominance of the media as theprimary information source for non-experts, whose inflation expectations show significant dispersioncompared to the well-anchored expectations of professionals. Recent episodes influenced by supplyshocks have further emphasized this disparity. While communications at ECB press conferences play acrucial role in clarifying policy intentions and reducing market uncertainty, their potential to shape theexpectations of non-experts is less understood. Non-experts tend to favor indirect communicationthrough media channels, suggesting that central banks may need to adapt their strategies to better engagethis audience and align their expectations with policy objectives. Direct communication could helpanchor citizens’ inflation expectations if it increases public attention to monetary policy newsandenhances the broad understanding of how monetary policy works. Identifying the effects of directcommunication on non-experts’ literacy and expectations in real-world settings is challenging, and afield experiment provides a practical approach to studying these effects. The main conclusion is thatdirect communication, similar to that of press conferences, shapes non-experts’ inflation expectationsby improving their knowledge of monetary policy and enhancing their trust. This conclusion is drawnfrom sessions with diverse international groups totaling 5,000 individuals who visited the ECB betweenDecember 2022 and May 2025 at its headquarters in Frankfurt and volunteered to participate. Experimental design and treatment: Visitors with varying levels of monetary policy knowledge attendedpolicy briefing sessions, akin to communication at ECB press conferences. We exposed them toinformation treatments and created exogenous variation in the information they received about monetarypolicy. Participants were randomly assigned by group session to one of three treatments: (i) Treatedgroup: it received presentations from ECB experts on the ECB’s monetary policy framework and itsdecisions; (ii) Control group: it received no treatment; (iii) Placebo group: it received presentations fromexperts about other ECB tasks, though unrelated to its monetary policy. We implemented a new surveyto gather insights into participants’ monetary policy knowledge, inflation, and growth expectationsbefore and after the treatment. The inclusion of a placebo group allowed us to prove that monetarypolicy-specific learning effects were genuine. The population primarily consisted of younger, financiallyeducated indiv